Journal of Clinical Apheresis 10S37-89 (1995)

Peripheral Blood Stem Cell Collection in a Patient With Chronic Myelogenous Leukemia

and a High Circulating Nucleated Red Cell Fraction

Jong-Hoon Lee, Charles C. Miraglia, William W. Grosh, and Paul D. Mintz

Department of Transfusion Medicine, Warren G. Magnuson Clinical Center, National Institutes of Health, Bethesda, Maryland (J.-H. L.), Departments of Pathology (C. C. M., P. D. M.) and Internal

Medicine (W. W. G., P. D. M.), University of Virginia Health Sciences Center, Charlottesville, Virginia

A high level of circulating nucleated red blood cells (NRBC) in patients with chronic myeloproliferative syndromes could potentially complicate peripheral blood stem cell (PSC) collection. The mononuclear NRBC might comprise a significant fraction of the total mononuclear cells in the final product. We report a successful PSC collection in a patient with more NRBC than WBC in the peripheral blood. A 27-year-old man with chronic myelogenous leukemia underwent eight PSC collection procedures, seven using the Cobe Spectra (Spectra) and one using the Fenwal CS3000 Plus (CS). PSC product manipulations to remove NRBC were unnecessq. As assessed by post-collection NRBC:WBC ratio as a percent of the initial ratio, Spectra selectively harvested mononuclear leukocytes over NRBC. The collected products had a mean NRBC:WBC ratio that was 3.4% of the peripheral blood ratio. Adequate numbers of mononuclear leukocytes were collected with less than 6% NRBC contamination. The single CS procedure resulted in a comparable NRBC reduction efficiency as the Spectra. We conclude that PSC harvest using automated blood cell separators from patients with a high level of circulating NRBC may result in a product with an acceptable number of NRBC. 0 1995 Wiley-Liss, Inc.

Key words: chronic myelogenous leukemia, nucleated red blood cells, peripheral blood stem cell collection

INTRODUCTION

An unusually high level of circulating nucleated red blood cells (NRBC) can be seen in patients with myloid metaplasia, myelofibrosis, evolving polycythemia Vera, essential thrombocythemia, and rarely, in patients with complicated chronic myelogenous leukemia (CML) [ 11. In such patients, a consideration of autologous blood stem cell (PSC) transplantation as a treatment option should include the anticipation of complications in har- vesting stem cells from the peripheral blood. Sedimenta- tion of NRBC under centrifugation within an automated blood cell separator could resemble that of mononuclear leukocytes [2,3]. An attempt at PSC collection using conventional blood cell separators and standard instru- ment settings could result in a product heavily contami- nated by NRBC. More procedures than anticipated might be required to harvest an adequate number of circulating hematopoietic progenitor cells. A large number of NRBC in the PSC products may necessitate product manipula- tion, complicate cryopreservation, or result in clinical complications at infusion. These considerations have not been addressed in the PSC transplantation literature. In this report, we describe our experience in harvesting PSC from a patient with CML and an unusually high level of circulating NRBC.

0 1995 Wiley-Liss, Inc.

MATERIALS AND METHODS

Patient

A 27-year-old man with Philadelphia chromosome pos- itive CML in chronic phase was referred for PSC collec- tion. A previous attempt at allogeneic bone marrow trans- plantation was unsuccessful (graft rejection), requiring autologous marrow rescue. Severe myelofibrosis excluded the possibility of a repeat autologous bone marrow har- vest. Alpha-interferon (IFN) was discontinued 1 day prior to the first PSC collection procedure in order to optimize PSC collection. The spleen had been removed previ- ously. The patient did not receive hematopoietic growth factor mobilization. The peripheral blood smear demon- strated red cell changes, including severe anisocytosis, poikilocytosis, and innumerable tear drop cells. Fifty-one NRBC per 100 leukocytes were present. Other blood counts were as follows: white blood cell (WBC) 21.1 X 109/L (corrected for NRBC; 75% neutrophils,

Received for publication March 14, 1993; revised February 28, 1995; accepted March 9, 1995.

Address reprint requests to Jong-Hoon Lee, Building 10, Room 1C- 71 1, National Institutes of Health, Bethesda, MD 20892-1 184.

88 Lee et al.

TABLE I. Instrument Settings and Related PSC Collection Parameters for Spectra and CS Blood Cell Separators*

Settings Spectra cs Whole blood flow rate (rnlimin) 59 60 Plasma pump rate (rnlhin) NR 47 Collection rate (rnlimin) 1 NA Mean product volume (rnl) I46 153 Interface setting 3% HCT 100 Chamber tvoe NA svcc CS program 8 location

Default setting

Modified setting

60 61 62 64 68 73 78

24 1

16 32 45 92 85

245 0

21 42 68

102 95

* NR = not recorded; NA = not applicable.

12% lymphocytes, 13% monocytes), hematocrit 26.7%, and platelets 323 X 1091L.

PSC Collection Eight liters of blood was processed for each procedure

using either Cobe Spectra (Spectra; Cobe BCT, Inc., Lakewood, CO) or Fenwal CS3000 Plus (CS; Baxter Corporation, Fenwal Division, Deerfield, IL) blood cell separator. A total of 8 procedures, 7 with Spectra and 1 with CS, were performed over 10 days (daily procedures excluding 2 days between collections 5 and 6). The in- strument settings and related PSC collection parameters used were those typical for harvesting PSC at our center, for both Spectra and CS blood cell separators. The inter- face setting on Spectra was maintained manually through- out each procedure by adjusting the collection rate so that the color of the collection line matches that for 3% HCT, as indicated on a standard color strip. For PSC collection using CS, program 8 default settings were modified at each indicated program location. The modified settings were typical for PSC collection procedures using CS at our center. The instrument settings and related collection parameters for both blood cell separators are detailed in Table I.

Cell Counts and Calculations NRBC and mononuclear leukocyte contents of the col-

lected products were assessed by standard manual cell differential counting. Product blasts were recognized based on morphologic criteria. The NRBC reduction efficiency was calculated by dividing the product NRBC:WBC ratio by the equivalent ratio in the peripheral blood. The mono-

TABLE 11. NRBC Reduction Efficiency of Spectra and CS*

NRBC/100 WBC NRBC reduction P BCS Peripheral Product efficiency (%)

1 S 51 2 4 2 S 65 I 2 3 S 88 4 5 4 S 96 4 4 5 S 157 4 3 6 S 225 6 3 7 C 230 6 3 8 S 225 6 3 Mean 142 4 3.4

*Mean values refer only to collections using Spectra. P = procedure; BCS = blood cell separator; WBC = white blood cell.

nuclear leukocyte collection efficiency was calculated by dividing the product yield by the number of cells pro- cessed.

RESULTS

Despite the high level of circulating NRBC, the PSC products were not significantly contaminated by NRBC. An NRBC reduction efficiency of 3.4%, with acceptable mononuclear leukocyte yield and collection efficiency, was achieved by both blood cell separators. Over 8 pro- cedures, a total of 14 X lo8 mononuclear leukocytesikg of patient weight (55 kg) were harvested. In comparison to 7 semi-automated Spectra procedures, the single more fully automated CS collection resulted in comparable NRBC reduction, a larger cell yield, more efficient mono- nuclear leukocyte collection, and a higher product hema- tocrit. Choosing an interface on Spectra that allows a higher product hematocrit may generate a product which more closely resembles a CS product. The NRBC reduc- tion efficiency and mononuclear leukocyte collection ef- ficiency are summarized in Tables I1 and 111, respec- tively.

DISCUSSION

The empirically selected apheresis schedule and blood cell separator settings resulted in the harvest of an accept- able number of mononuclear leukocytes with minimal NRBC contamination. Spectra and CS performed equally well in selectively harvesting mononuclear leukocytes over NRBC with less than 6% NRBC contamination, and the NRBC:WBC ratio in the PSC product was approxi- mately 3% of the peripheral ratio. In comparison, hy- droxyethyl starch separation [4] yielded a final NRBC: WBC ratio which was 16% of the peripheral ratio, and Ficoll-Hypaque method [5] reduced the ratio to 23% of the initial ratio (data not shown). The cell yield and the

Peripheral Stem Cell Collection and NRBC 89

are clonal processes affecting multiple hematopoietic cell lines [ 11. In the current case, the discontinuation of IFN therapy probably resulted in a kinetically coupled in- crease in the production of NRBC and WBC, which suggests that both cell lines are derived from a common CML precursor. The progressive increase in the product blast content (Table 111) also supports increased cell pro- duction as the result of a clonal process. The patient’s presumed extramedullary hematopoiesis is unlikely to be a physiologically adaptive consequence of marrow fibro- sis or demands in oxygen delivery.

TABLE 111. Mononuclear Leukocyte Yield and Collection Efticiencv*

P

1 2 3 4 5 6 7 8

- BCS

S S S S S S C S

PWBC

21 22 21 22 19 24 26 30

Yield CE

0.99 12 1.1 I 1 1 .5 20 1.1 12 I .o 17 3. I 35 4.1 53 1 . 1 15

Product Product HCT blasts

4 2 4 1 4 5 3 3 2 5 6 8

13 14 3 14

* P = procedure; BCS = blood cell separator; PWBC = peripheral white blood count ( X IO’/L); Yield = mononuclear leukocyte yield ( X lo8 cells/kg); CE = collection efficiency (yield divided by the num- ber of mononuclear leukocytes processed, percent); HCT = hematocrit (percent).

hematocrit in the PSC products were as expected based on our previous experience for both blood cell separators.

The remarkable efficiency of both blood cell separa- tors in selectively harvesting mononuclear leukocytes over NRBC is an unexpected, surprising result in view of the morphologic similarity of the two cell types; cell sedi- mentation may be also similar. Our results suggest that the sedimentation characteristics of NRBC within a cen- trifugal separation chamber more closely resemble those of mature red cells than those of mononuclear leukocytes. Cell separation within a centrifugal force field depends on the relative cell sedimentation rates, and cell sedimen- tation in turn is affected by many variables, including cell density, size, defonnability, surface charge, tendency to aggregate, and the nature of the suspending medium [2,6,7]. Despite our successful experience, small changes in these variables may have a major impact on the overall efficiency in NRBC reduction and in mononuclear leuko- cyte collection. Further clinical experience is necessary in order to predict accurately the outcomes of PSC har- vests in patients with a high level of circulating NRBC.

The peripheral NRBC:WBC ratio depends on multiple factors, including cell production rates, removal or de- struction rates, and cell compartmentalization for both NRBC and WBC. The accelerating increase in the pe- ripheral NRBC:WBC ratio (Table 11) probably reflects both increasing cell proliferation upon discontinuation of IFN and direct effects of PSC collection. The WBC re- moval itself may induce alterations in cellular kinetics independent from dicontinuation of IFN therapy, and the NRBC:WBC ratio could have been influenced in the short term by the selective WBC removal over NRBC. Al- though the effect of PSC harvest on hematopoietic pro- genitor cell proliferative activity in CML remains un- clear, the concomitant increase in the number of WBC and NRBC in the peripheral blood supports the view that CML and related chronic myeloproliferative disorders

CONCLUSIONS

The current case report demonstrates that a high level of circulating NRBC does not necessarily complicate PSC collection, and that special PSC product manipulations to reduce NRBC may not be necessary. Further experience is necessary in order to characterize further the role of NRBC in harvesting mononuclear leukocytes using auto- mated blood cell separators.

ACKNOWLEDGMENTS

The authors sincerely thank Cathy Ciucias and Leslee Lindquist for skillful hemapheresis operation; Sarah Don- nelly, Anne Hobson, Gerda Pirsch, and Mitsi Wood for excellent technical assistance; and Charles Carter for help- ful technical information.

REFERENCES

1. Lichtman JA: Chronic myelogenous leukemia and related disor- ders. In Williams WJ, Beutler E, Erslev AJ, and Lichtman MA (eds): “Hematology.” New York: McGraw-Hill Publishing Com- pany, 1990, pp 193-235.

2. Brown RI: The physics of continuous flow centrifugal cell separa- tion. Artif Organs 12:4-20, 1989.

3. Shortnian K , Haskill JS, Szenberg A, Legge DG: Density distribu- tion analysis of lymphocyte populations. Nature 216:1227-1228, 1967.

4. Warkentin P, Hilden JM, Kersey JH, Ramsay NKC, McCullough J: Transplantation of major ABO-incompatible bone marrow depleted of red cells by hydoxyethyl starch. Vox Sang 48:89-104, 1985.

5. Wells JR, Sullivan A, Cline MJ: A technique for the separation and cryopreservation of nyeloid stem cells from human bone marrow. Cryobiology 16:201-210, 1979.

6. Van Wie BJ, Sofer SS: Sedimentation theory and practical consid- erations for the design of centrifugal blood cell processes. Int J Artif Org 7:215-222, 1984.

7. Chien S , King RG, Skalak R, Usami S, Copley AL: Viscoelastic properties of human blood and red cell suspensions. Biorheology 12:341-346, 1975.